|Publication number||US7602403 B2|
|Application number||US 11/216,345|
|Publication date||Oct 13, 2009|
|Filing date||Aug 30, 2005|
|Priority date||May 17, 2001|
|Also published as||US7126578, US20060001679|
|Publication number||11216345, 216345, US 7602403 B2, US 7602403B2, US-B2-7602403, US7602403 B2, US7602403B2|
|Original Assignee||Adobe Systems Incorporated|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (22), Non-Patent Citations (2), Referenced by (3), Classifications (13), Legal Events (2)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application is a continuation and claims the benefit of priority under 35 U.S.C. Section 120 of U.S. application Ser. No. 10/098,328, filed Mar. 15, 2002, now U.S. Pat. No. 7,126,578 which claims priority to U.S. Provisional Application Ser. No. 60/292,193, filed May 17, 2001. The disclosure of the prior applications is considered part of and is incorporated by reference in the disclosure of this application.
The present invention relates to combining raster and vector data in the presence of transparency.
Generally, graphics computer programs such as the Photoshop® available from Adobe Systems Incorporated of San Jose, Calif., have imaging models that use both raster and vector data. These imaging models generally have constructs that support compositing effects such as layering and opacity, including per pixel transparency with respect to raster data, object level transparency, and color blending.
Some imaging models cannot support some of the compositing effects mentioned. Specifically, these imaging models lack the constructs needed to support the compositing effects. These imaging models will be referred to as simple imaging models.
In order to display a composition created from a first program having the mentioned constructs in a second program that lacks these constructs, i.e., one that has a simple imaging model, it is necessary to map data of the composition into a format compatible with the second program. Typically, the mapping causes a loss of vector data. Because of this loss, an image of the composition rendered by the second program sometimes has pixelated edges and appears fuzzy. An image that appears fuzzy will be referred to as a fuzzy image. An image that does not appear fuzzy will be referred to as a crisp image.
The present invention provides methods and apparatus, including computer program products, for combining raster and vector data into a format that preserves the resolution-independent quality of vectors.
In one implementation, a system in accordance with the invention combines raster and vector data in three phases: an analysis phase, an optional sorting phase, and an output phase. In the analysis phase, the system builds a list of vector objects where each object has a sequence of clipping states. Generally, a clipping state of an object is an appearance of the object within a certain region. Associated with each clipping state are clipping instructions and instructions on how to render the portion of the object clipped or the entire object if there is no clipping path. Clipping states are further described below. In the optional sorting phase, the system sorts the list of vector objects so it can avoid processing vector objects or portions of vector objects that are not needed to render an image because the vector objects or their portions are concealed. In the output stage, the system recursively processes the list of vector objects to generate an output image.
In general, in one aspect, the invention provides a method for preparing one or more vector data objects for rendering. The method includes receiving a first object that is a vector data object. The method includes defining clipping states of the first object. Each clipping state of the first object represents an appearance of the first object for a region of the first object. The method includes determining, for each clipping state, whether a clipping path is needed to render the appearance of the first object represented by the clipping state. When a clipping state is needed, the method includes generating clipping instructions for calculating a clipping path for the clipping state. When a clipping path is not needed, the method includes generating clipping instructions indicating that a clipping path is not needed. The method includes generating, for each clipping state, rendering instructions for rendering the appearance of the region of the first object represented by the clipping state. The method includes associating the generated clipping and rendering instructions to the corresponding clipping states.
The invention can be implemented to realize one or more of the following advantages. Raster and vector data can be combined into a format that preserves the resolution-independent quality of vectors and that is fully supported by a simple imaging model. The raster and vector data combination preserves data describing per pixel transparency, object level transparency and color blending effects, and computed effects such as layering. A system in accordance with the invention supports multipart vector objects without having to convert them into general artwork. The system suppresses superfluous branches of recursions when recursively clipping. The system applies the technique of object concealment to cases where an object is partially concealed. The system needs information about imaging models only at the analysis phase and not at the optional sort or the output phases. Consequently, the recursive portion of the output phase does not need to change if the imaging model changes. The recursion portion is also simplified by not trying to include optimizations for details of the imaging model. The system limits deep recursions and saves computing resources.
Like reference symbols in the various drawings indicate like elements.
Methods and apparatus in accordance with the invention map imaging models of programs such as the Photoshop program into a format fully supported by programs that use simple imaging models such as the PostScript® and portable document format (“PDF”) imaging models. Note that the techniques described below are not specific to the Photoshop imaging model and apply to other imaging models. For example, the techniques can be used to map PDF version 1.4 images back to PDF version 1.3 images.
The output of conversion processes in accordance with the invention that map one imaging model into another generally includes some indication of a region to clip to (using vector clipping) and a raster with which to fill that region. Where possible, the conversion process also retains elements as pure vector artwork such as filled or stroked paths and text. The conversion process recognizes that other constructs such as solid colors or gradients can be used in place of rasters.
As shown in
The PostScript imaging model allows a user to draw these three items in succession. The person 110 can be masked by a clipping path. However, a clipping path cannot adequately define the way the person's hair, which is blowing in the breeze, should interact with other elements in the image. While a clipping path allows a user to specify which portions of the image are part of the person and which portions are not, a clipping path does not allow a user to specify that the hair should be partially transparent in its thinnest areas. Using a clipping path would produce an image having a hard and unnatural transition between the person's hair and the background.
A hard transition is exactly what is desirable, however, around the edges of the type 120. One approach to building image 100 is to create the composition in a program like Photoshop and simply generate a raster image. If one does so, however, the edges of the type 120 will not be rendered at the maximum possible resolution of an output device such as an image setter unless the entire raster image is also generated at that resolution. Rendering the entire image at the maximum resolution results in a huge composition file. Rendering the entire image at a resolution other than the maximum resolution results in pixelated edges. Thus, with the approach of compositing in Photoshop® and generating a raster image, a user ends up with either pixelated edges on the type or huge files.
Note that the described technique can also handle cases where the text is partially opaque. In these cases, the image for the text interior shows the interaction of the text color with the background in addition to including the person.
The described technique illustrates two methods of conversion, recursive clipping and planar mapping. In each of these methods of conversion, vector data is expressed as clipping data. The output image is then calculated for each clipping path.
In both recursive clipping and planar mapping, image 300 is generated from four components: background 310 alone, background 310 plus the content of circle 320, background 310 plus the content of circle 330, and background 310 plus the contents of both circles 320 and 330. The content includes color data and blending instructions inside the circle, ignoring the boundary of the circle. The distinction between recursive clipping and planar mapping is in the clipping paths used.
Generally, recursive clipping renders a composition by building and then processing subsets of a set of vector objects. For each subset, a clipping path is defined from the outlines of each element in the subset. Then, an image is drawn based on including the corresponding graphic elements and excluding the elements not represented in the subset. Furthermore, the clipping process can be simplified in a program, such as a PostScript interpreter, that supports pushing and popping a clipping stack with the current total clip being defined as the intersection of all of the clips on the stack.
Draw background 310 (item 410)
Push a clip for circle 320
Draw background 310 plus the content of circle 320 (item 420)
Pop the clip for circle 320
Push a clip for circle 330
Draw background 310 plus the content of circle 330 (item 430)
Push a clip for circle 320
Draw background 310 plus the content of both circles 320,
330 (item 440)
Pop the clip for circle 320
Pop the clip for circle 330
This approach is an application of a standard method for enumerating subsets of a set. It is also the format generated by the method of artwork filters discussed below.
Recursive clipping has the benefit of leaving all of the outline intersection calculations to a raster image processor (“RIP”). Having the RIP calculate outline intersection allows the calculations to be performed at the resolution of the RIP. Having the RIP calculate outline intersections also places the original outlines in an output file. In particular, if a format supports clipping to text, then the text can be placed in the output file without further computation of outlines.
Recursive clipping also has the benefit that an anti-aliasing rasterizer does not have to do subpixel sampling in order to avoid halos or gaps at the edges. This is because the recursive approach does not rely on the detection of abutting edges by the rasterizer.
There are two significant disadvantages to recursive clipping. First, recursive clipping requires the RIP to support a potentially deep set of nested clips. Falling back to rasterized vectors when a threshold for active clips is exceeded mitigates this problem. The other disadvantage is that the power set of the set of vector objects grows exponentially as the number of objects grows. Hence, to make the recursive clipping approach practical, it is necessary to aggressively reduce the set of subsets considered. Reducing the number of subsets will be referred to as pruning. The simplest approach to pruning is to note that once the intersection of the clip regions becomes empty, it is possible to discard that subset and any subset containing that subset. Because it is advantageous to avoid processing all of the geometry in detail and it is not possible to account for the exact scan conversion behavior of the final RIP, a conservative approximation to the area covered by a clip region is necessary. The simplest such approximation is a bounding box.
The pseudo-code shown in Table 2 below represents one recursive clipping method. The code generates a series of clipping operations and image drawing operations. Vector objects can be set to one of three states when rendering: hidden, shown, and content-only. The hidden state causes an object to not participate in the compositing process. The shown state causes an object to participate just as it normally would. The content-only state causes an object to participate normally except that the outline of the object is not considered. That is, the object's content is not clipped by the object's path. All of the vector objects start out in the shown state. All non-vector objects, such as rasterized objects, are always set to be in the shown state. The pseudo-code below is applied to a composition bounded by a rectangle. The composition also includes a set of vector objects. The recursive clipping process is started by invoking the pseudo-code with a list of all vector objects in the set and the rectangle bounding the composition. The pseudo-code composites the image for a region of interest (the rectangle, in this case).
(objects: LIST OF VectorObject; rect: Rectangle);
IF EmptyList (objects)THEN
LET obj: VectorObject :=First (objects);
LET objClip: VectorData :=GetVectorData (obj);
LET objBounds: Rectangle :=VectorDataBounds (objClip);
LET overlap: Rectangle :=Intersect (rect, objBounds);
SetVectorObjectRenderingState (obj, hidden);
GenerateRecursiveClippingOutput (Rest (objects), rect);
IF NOT EmptyRectangle (overlap)THEN
SetVectorObjectRenderingState (obj, contentOnly);
(Rest (objects), overlap);
OutputPopClip ( );
SetVectorObjectRenderingState (obj, shown);
The implementation of image scaling in PostScript, PDF, and the lossy compression scheme used in Joint Photographic Experts Group (“JPEG”) can cause stitching artifacts in any output that relies on rendering images inside of clipped regions. These artifacts will show up as areas where the colors output by the RIP on either side of a clipping path should match but fail to do so.
The PDF and PostScript scaling problems arise from the fact that drawing one big image does not necessarily result in the same thing as drawing the image in multiple pieces. To avoid stitching artifacts, it is necessary to draw an image that is the size of the entire artwork. However, most of this image would never be shown because only the portions inside the clipping is needed. The remaining portions can be represented in as compressed a form as possible. It may also be necessary to pad the region of interest to account for the effects of lossy compression schemes such as that of the JPEG.
One additional item to note with regard to recursive clipping is that while the size of the output file may grow exponentially with the number of objects if pruning does not work, the space consumed in the computation is linear in the number of objects. More sophisticated representations for the approximation to the current total clip can also result in an increase in space, but this is dependent on the precision desired and not on the number of objects.
In contrast to recursive clipping, planar mapping does not look at the subsets of the set of vector objects. Instead, it looks at the regions into which the set of vector objects carve a plane of the composition. In principle, each pixel is associated with exactly one region once the plane of the composition plane is carved. In practice, it is necessary to represent pixels along boundaries in the images for each region.
Generally, planar mapping separates an image into atomic regions. The basic procedure for planar mapping is to take each vector object in turn and add it to an accumulated planar map. Adding it to the map splits some regions into two. Associated with each region is a list of vector objects whose outlines bound the region. When the planar mapping is complete, the output image of the composition is built by processing each region, using the associated list of objects to render the portion of the image within the region.
Planar mapping trades the previously discussed potentially explosive growth of output size for the complexity of building a planar map. The program for planar mapping is much more complicated than that for recursive clipping. Additionally, the planar mapping approach consumes more memory space than does the recursive clipping approach. If vector objects are defined by Bézier curves, then the maximum number of regions in the map varies as the square of the number of Bézier segments in the vector objects. For objects, such as types, that have outlines of complex shape, the number of regions resulting from planar mapping increases significantly. The increase is not an exponential function of vector objects but rather a quadratic function of the number of Bézier segments.
Associated with each planar region is a list of vector objects. At worst, the memory space consumption grows approximately proportional to the cube of the number of vector objects. Subdividing the image into smaller rectangles and excluding objects outside those rectangles mitigates this memory space consumption problem. But if a region has too many complex outlines, the subdividing solution does not always avoid the memory space consumption problem. This problem too can be mitigated. If lists of both included and excluded objects are kept for each region, subdivision can be limited and rasterized vectors can be used when either or both of these lists reached a size limit.
Planar mapping puts anti-aliasing rasterizers at a disadvantage because it requires the rasterizer either to use super-sampling or to recognize the shared edges on independent objects. If this is a significant concern, a format designed specifically to represent planar maps may be needed. The PostScript and PDF models do not provide such formats. A special format for planar maps can also lead to faster rasterization.
The following paragraphs describe artwork filters. Some vector-based illustration programs support the notion of filling a shape with a filter that operates on underlying artwork. For example, a circle can be filled with a magnification effect that results in an enlarged version of a portion of the underlying artwork to be seen through the circle. Transparency can be treated as a filtering effect by viewing the compositing of a color or colors with the underlying artwork as a color-transforming filter on the artwork. Generating an output from this method includes the following steps. An output stream for the elements beneath a filter object is generated. This stream is outputted. The filter is applied to the output stream to produce a filtered output stream. Instructions to clip to the object are outputted. The filtered stream is outputted. Instructions to restore the clip are outputted. The above steps do not apply only to output streams. They also apply to other representations such as artwork trees. When present, multiple filter objects are processed recursively using a principle similar to that of recursive clipping. The difference between the two principles is that instead of taking an execution path for each subset, the build up is done in the concatenation of the output streams. Alternatively, a stack of active artwork filters is maintained. This alternative more closely mirrors the principle of recursive clipping.
One advantage of the artwork-filter method is that the method can handle spatial transformation effects inside shapes. Both the recursive clipping and the planar mapping approaches cannot handle spatial transformation effects well because they are both concerned with finding the shapes that overlap particular pixels. On the other hand, the artwork-filter method potentially processes a significant amount of data to get its job done. The size of this data can grow exponentially.
The artwork-filter method also requires that everything interesting be treated as a filter. For example, to handle the initial example, one needs a filter object type that deals with raster data containing transparency.
In one implementation, a system in accordance with the invention performs an augmented recursive-clipping method. The augmentation includes the improvements listed below.
The first two enhancements allow the system to handle computed effects and to provide vector output for some of the effects. The second two enhancements provide significant improvements in the process of pruning the number of subsets the system must consider. The first enhancement also prunes the number of combinations of a vector object that must be considered. Without clipping states, it is necessary to represent any object with a sequence of more than two clipping states as multiple objects. For example, a vector object having a stroke and an interior must be represented as a stroke object and an interior object. The two objects require four combinations rather than three to examine in a recursive pass.
As shown in
As shown in
A state representing the exterior of the masking object in a clipping group conceals everything within the group. Note that to be included in a set of concealed objects, it is not necessary to have complete concealment of the object by the area bounded by the region of the clipping state. All that is required is that a pixel within the region of the clipping state be not visible because of attributes of the state and the state's role within the composition. The system represents the set of concealed objects by a general set construct. Alternatively, the system can represent the set of concealed objects with a range of object indices.
For each state the system uses the desired clip to calculate a “may cover” region and a “must cover” region (step 750). The may cover region corresponds to any locations that might get marked by a RIP when using the current clip. The must cover region corresponds to any locations that must fall inside the current clip when processed by the RIP. One can view these as being calculated by taking the desired clip and expanding and contracting it respectively for scan conversion precision issues. The calculation does not depend on a particular representation for these regions. It depends on being able to perform intersections and subtractions with the regions, to test the regions for emptiness, and to get a bounding box around the regions. Possible data representations include run arrays or bitmaps at a particular resolution. One implementation of the system uses a combination of a bounding rectangle and a bitmap. Optionally, the system defines the regions analytically. However, path algebra can be relatively complicated and it can be difficult to avoid creeping precision problems with repeated processing.
For each clipping state currently being considered, the system iterates through subsequent clipping states, i.e., states that are processed after the current state is processed, and subtracts the must cover region of the subsequent clipping state from the may cover region of the current clipping state (step 760). At the output stage, the pixels in a must cover region of a subsequent state will get marked eventually and hence do not need to be marked by the earlier state. The system prunes its recursion in the output stage when the intersection of may regions of active clipping states become empty (i.e., the may regions do not intersect). By reducing the area covered by the may regions, the intersections of the may regions become empty more frequently. Hence, the system can prune aggressively. At the end of the analysis phase for an individual vector object, the system discards the must cover regions for the clipping states of the vector object because these must cover regions are no longer needed. Depending on the exact sequence of operations used to generate the clipping states and to subtract must regions of subsequent states from may regions of a current state, the system can dispose of the must regions before the end of the analysis phase.
Optionally, if the system generates the clipping states in order, the system can actually do the region calculations and subtractions as it accumulates the list of states for an object.
The described may-must-mark calculation allows the system to bypass processing portions of an image that do not appear in the final output image. For example, see
Referring back to
The system generates a final output image (step 630). In addition to maintaining the previous information, the system maintains a concealment count per vector object. These counts are initialized to zero. The pseudo-code in Table 3 below shows one method for generating the final output image.
(objects: LIST OF VectorObject; mayMark: Region);
IF EmptyList (objects)THEN
LET obj: VectorObject :=First (objects);
IF 0 <GetConcealmentCount (obj)THEN
GenerateOutput (Rest (objects), mayMark)
FOR i: Integer :=0 TO GetClipStateCount (obj) − 1 DO
LET state: ClipState :=GetClipState (obj, i);
LET stateMayMark: Region
LET overlap: Region :=
IF NOT EmptyRegion (overlap)THEN
LET concealsCount: Integer :=
FOR j: Integer :=0 TO concealsCount − 1
(GetConcealedObject (state, j))
GenerateOutput (Rest (objects), overlap);
OutputPopClip ( );
FOR k: Integer :=0 TO concealsCount − 1
It is possible to elide pop clip operations with push clip operations. In particular, for unclipped clip states, e.g., the exterior of vector objects, the implementation suppresses the pushing and popping of the clip data entirely.
Similar to the method described in reference to Table 1 for performing recursive clipping, to start an augmented recursive clipping process, the system calls a routine implementing the pseudo-code of Table 2 and inputs a sorted-object list and a region that represents the entire composition.
Optionally, the system limits the clipping depth. The system counts the number of times it has clipped. When it has exceeded the depth limit, the system invokes OutputImageForCompositedArtwork just as if the object list had become empty.
Another option is to pass the may-mark region to OutputImageForCompositedArtwork instead of passing just the region's bounding box. This option limits the compositing further or can be used to increase the compression on the hidden areas outside the region when writing the image to the output. The compression increase results in an output with pieces that looked like crude approximations to the results from planar mapping.
In summary, an augmented recursive clipping method provides the following features. The system can handle multi-part vector objects without having to convert them into general artwork. Note that one form of a multi-part vector object is a planar map. Hence, the system encompasses artwork that has planar-mapped objects though it does not build a planar map for the totality of the objects.
By accounting for other areas that will be marked later, the may-must-mark calculations prune many superfluous branches of the recursion. In particular, if one vector object is completely contained within the bounds of another object, then the system recognizes that the exterior of the outer object intersected with the interior of the inner object is empty. The conventional recursive clipping process does not detect this problem.
The augmented recursive method incorporates the notion of vector objects concealing other objects in a general way. That is, the method allows the exterior of an object to conceal other objects.
The augmented recursive method isolates all understanding about the actual imaging model into the analysis phase. The system does not have to consider the imaging model in the optional sorting or the outputting phase, in which the system performs the recursions. The actual recursive process can work with a variety of imaging models.
The recursion-depth limits provide a useful fallback to prevent the results from growing too explosively as a result of a deep recursion.
The invention can be implemented in digital electronic circuitry, or in computer hardware, firmware, software, or in combinations of them. Apparatus of the invention can be implemented in a computer program product tangibly embodied in a machine-readable storage device for execution by a programmable processor; and method steps of the invention can be performed by a programmable processor executing a program of instructions to perform functions of the invention by operating on input data and generating output. The invention can be implemented advantageously in one or more computer programs that are executable on a programmable system including at least one programmable processor coupled to receive data and instructions from, and to transmit data and instructions to, a data storage system, at least one input device, and at least one output device. Each computer program can be implemented in a high-level procedural or object-oriented programming language, or in assembly or machine language if desired; and in any case, the language can be a compiled or interpreted language. Suitable processors include, by way of example, both general and special purpose microprocessors. Generally, a processor will receive instructions and data from a read-only memory and/or a random access memory. The essential elements of a computer are a processor for executing instructions and a memory. Generally, a computer will include one or more mass storage devices for storing data files; such devices include magnetic disks, such as internal hard disks and removable disks; magneto-optical disks; and optical disks. Storage devices suitable for tangibly embodying computer program instructions and data include all forms of non-volatile memory, including by way of example semiconductor memory devices, such as EPROM, EEPROM, and flash memory devices; magnetic disks such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM disks. Any of the foregoing can be supplemented by, or incorporated in, ASICs (application-specific integrated circuits).
To provide for interaction with a user, the invention can be implemented on a computer system having a display device such as a monitor or LCD screen for displaying information to the user and a keyboard and a pointing device such as a mouse or a trackball by which the user can provide input to the computer system. The computer system can be programmed to provide a graphical user interface through which computer programs interact with users.
The invention has been described in terms of particular embodiments. Other embodiments are within the scope of the invention. For example, steps of the invention can be performed in a different order and still achieve desirable results.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3874978 *||Nov 10, 1972||Apr 1, 1975||Visual Graphics Corp||Shadow lettering system|
|US4408198 *||Sep 14, 1981||Oct 4, 1983||Shintron Company, Inc.||Video character generator|
|US5081594 *||Jan 31, 1989||Jan 14, 1992||Kroy Inc.||Real-time rasterization system for a column-oriented printing apparatus or the like|
|US5206718 *||Aug 17, 1990||Apr 27, 1993||Ricoh Company, Ltd.||Color image processor for creating shadows and masks|
|US5396585||Jan 24, 1992||Mar 7, 1995||Ricoh Company, Ltd.||Clipping processor|
|US5555353 *||Jul 23, 1993||Sep 10, 1996||Dainippon Screen Manufacturing Co., Ltd.||Method of and apparatus for producing shadowed images|
|US5579416 *||May 23, 1994||Nov 26, 1996||Canon Kabushiki Kaisha||Character processing apparatus for selectively modifying a font pattern|
|US5651107||Aug 16, 1994||Jul 22, 1997||Sun Microsystems, Inc.||Method and apparatus for presenting information in a display system using transparent windows|
|US5724494 *||Jul 21, 1995||Mar 3, 1998||Canon Information Systems Research Australia Pty Ltd||Optimization method for the efficient production of images|
|US5777625||Mar 28, 1996||Jul 7, 1998||Hewlett-Packard Company||Clipper re-uses storage locations allocated for verticles|
|US5923381||Nov 25, 1996||Jul 13, 1999||Thomson Broadcast Systems||Device and method for processing a signal with a subject moving against a colored background|
|US5966136 *||Apr 12, 1995||Oct 12, 1999||Hewlett-Packard Co.||Efficient method for clipping numerous objects against an arbitrary clipping path|
|US6005586||Dec 12, 1997||Dec 21, 1999||Fuji Xerox Co., Ltd.||Drawing processing apparatus|
|US6014147 *||Nov 19, 1997||Jan 11, 2000||Canon Information Systems Research Australia Pty Ltd||Computer machine architecture for creating images from graphical elements and a method of operating the architecture|
|US6069633 *||Sep 18, 1997||May 30, 2000||Netscape Communications Corporation||Sprite engine|
|US6236410 *||Mar 15, 1999||May 22, 2001||Canon Kabushiki Kaisha||Efficient methods for the evaluation of a graphical programming language|
|US6369830 *||May 10, 1999||Apr 9, 2002||Apple Computer, Inc.||Rendering translucent layers in a display system|
|US6377269||May 28, 1999||Apr 23, 2002||Robert L. Kay||Automated generation of masks for photo-compositing|
|US6529208||Jan 6, 2000||Mar 4, 2003||International Business Machines Corporation||Method and apparatus for updating a window identification buffer in a data processing system|
|US6535635||Oct 25, 1999||Mar 18, 2003||Xerox Corporation||Error clipping for vector error diffusion|
|US6746331 *||Jul 14, 2000||Jun 8, 2004||Kabushiki Kaisha Sega Enterprises||Assisting a player to contact an object by adjusting timing|
|US6867787 *||Mar 13, 2000||Mar 15, 2005||Sony Corporation||Character generator and character generating method|
|1||Adobe(C) Photoshop(C) 6.0 User Guide for Windows(C) and Macintosh, Chapters 2 and 8, (C) 2000 Adobe Systems Incorporated, 62 pages.|
|2||Rose, Carla, "Sams Teach Yourself Adobe Photoshop 5.5 in 24 Hours", Sams Publishing, Copyright 2000.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US9317943 *||Oct 7, 2013||Apr 19, 2016||Oracle International Corporation||Interactive intersection areas|
|US20080082549 *||Oct 2, 2007||Apr 3, 2008||Vic Baker||Multi-Dimensional Web-Enabled Data Viewer|
|US20150029215 *||Oct 7, 2013||Jan 29, 2015||Oracle International Corporation||Interactive intersection areas|
|U.S. Classification||345/625, 345/628, 345/469.1, 345/626, 345/627, 345/468, 345/467|
|International Classification||G09G5/00, G06T11/00|
|Cooperative Classification||G06T11/60, G06T15/503|
|European Classification||G06T15/50B, G06T11/60|
|Aug 30, 2005||AS||Assignment|
Owner name: ADOBE SYSTEMS INCORPORATED, CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:HAMBURG, MARK;REEL/FRAME:016948/0330
Effective date: 20020315
|Mar 6, 2013||FPAY||Fee payment|
Year of fee payment: 4